The study reports novel findings on the associations of muscle weakness and atrophy with a differential expression profile of epigenetic events in the quadriceps of patients with advanced COPD, which was not seen in patients without muscle weakness.

CLINICAL PERSPECTIVES

  • Limb muscle weakness and decreased muscle mass are reliable predictors of COPD mortality, regardless of the severity of the respiratory condition. Whether epigenetic events involved in the control of muscle mass and function may be differentially expressed in patients with severe COPD and muscle weakness remains unknown.

  • Non-muscle-weakness and muscle-weakness COPD patients and healthy controls were studied. Compared with controls, in limb muscles of muscle-weakness patients, expression of muscle-enriched miR-1, miR-206 and miR-27a, levels of lysine-acetylated proteins and histones and acetylated histone 3 were increased; although expression of HDAC3, HDAC4, SIRT-1, IGF-1 were decreased, Akt expression did not differ, follistatin expression was greater, whereas myostatin expression was lower, SRF expression was increased and fibre size of fast-twitch fibres was significantly reduced.

  • These findings may help design novel therapeutic strategies (enhancers of miRNAs promoting myogenesis and acetylation inhibitors) to selectively target muscle weakness and atrophy in severe COPD patients.

INTRODUCTION

Chronic obstructive pulmonary disease (COPD) is currently considered to be a major cause of death worldwide [1,2]. In COPD, systemic manifestations and comorbidities are characteristic features that clearly have a negative effect on the patients' exercise capacity and quality of life [3,4]. Indeed, quadriceps muscle dysfunction was shown to occur in one-third of the patients with COPD even at early stages of their disease [5]. Furthermore, limb muscle weakness and decreased muscle mass are known to be important reliable predictors of COPD mortality, regardless of the severity of the respiratory condition [6,7].

Hypoxia, hypercapnia and acidosis, exacerbations, nutritional abnormalities, cigarette smoke and sarcopenia (aging) are part of the aetiology of COPD muscle dysfunction [8,9]. Biological mechanisms [9], such as structural abnormalities [10], muscle remodelling [11], oxidative stress [12] and metabolic derangements [13], are also important contributing factors to muscle dysfunction in COPD. Epigenetic control of cells, described as the process whereby gene expression is regulated by heritable mechanisms that do not affect DNA sequence, has recently emerged as a relevant contributor to muscle adaptation to environmental factors such as immobilization, exercise, muscle mass maintenance and disuse muscle atrophy models [1418]. For instance, spaceflight and hind-limb suspension experimental models induced a down-regulation of non-coding ssRNA molecules (miRNAs) such as miR-206 in skeletal muscles [19]. Interestingly, muscle overload up-regulated miR-206, whereas it down-regulated miR-1 and miR-133 expression in the same animals [17].

In COPD patients with preserved body composition, a pioneering study [16] demonstrated that miR-1 levels were reduced, whereas those of histone deacetylase (HDAC)4 were increased in their vastus lateralis (VL). The same authors also found [20] that levels of the transcription factor Yin Yang (YY)1, which modifies histones, inversely correlated with the size of slow- and fast-twitch fibres in the limb muscles of COPD patients with normal body composition. Furthermore, plasma levels of muscle-specific miRNAs were also shown to be increased in patients with severe COPD [15]. Recent investigations have also shown an increase in miR-1 expression in the VL of patients with mild COPD [21], whereas a reduction in muscle-specific miRNAs was detected in the diaphragm of patients with moderate-to-severe COPD [22]. Histone acetylation, defined as the balance between histone acetyltransferases (HATs) and HDACs was also shown to regulate muscle plasticity and mass in response to environmental factors in experimental models [23,24]. Despite this knowledge, it remains to be elucidated whether epigenetic events involved in the control of muscle mass and function may be differentially expressed in patients with severe COPD and muscle weakness. Interestingly, experimental models have also shown that small ubiquitin-related modifier (SUMO) ligases may favour sarcopenia and muscle wasting through premature senescence of satellite cells [25]. Whether this mechanism could also underlie muscle weakness in COPD remains unknown.

Therefore, our hypothesis was to explore whether quadriceps muscle weakness may be associated with a differential expression profile of epigenetic events so far demonstrated to be involved in muscle development and maintenance in patients with advanced COPD. Moreover, downstream pathways of muscle growth and atrophy shown to be regulated by those epigenetic events have also been analysed in the current study. Accordingly, our primary objectives were: (1) to explore expression levels of muscle-enriched miRNAs and their downstream targets involved in muscle growth and atrophy signalling, histone acetylation, HATs, HDACs, myogenic factors, including the myocardin-related transcription factor (MRTF)–serum response factor (SRF) axis and SUMO-2/3 in the quadriceps of patients with advanced COPD with and without muscle weakness and (2) to analyse potential correlations between the different study variables among the patients. Specimens from VL were additionally obtained in sedentary healthy controls for the purpose of the investigation and all groups of patients and control individuals were clinically and functionally evaluated.

METHODS

(See the Supplementary Material for detailed information on all the study methodologies).

Study subjects

A total of 41 male patients with stable advanced COPD [1,2,8,26] and 19 age-matched male sedentary controls of similar smoking history were recruited. Specimens from the VL were obtained from all subjects on an out-patient basis (COPD clinic). Additionally, COPD patients were further subdivided into those with and without muscle weakness (n=25 and n=16 respectively) according to a previous report on the prevalence and degree of muscle weakness in a European-based study [5]. In this study [5], one-third of the COPD patients exhibited muscle weakness, which consisted of a 25% reduction in quadriceps force compared with that observed in healthy controls. On this basis, we established 29.25 kg as the cut-off value to subdivide the patients in the current study. In a post-hoc analysis, patients with and without muscle weakness were also compared with healthy controls. Smoking history was similar between patients and healthy controls. The current cross-sectional investigation was designed in accordance with both the ethical standards on human experimentation in our institutions and the World Medical Association guidelines (Helsinki Declaration of 2008) for research on human beings. Approval was obtained from the institutional Ethics Committees on Human Investigation (Hospital del Mar and Hospital Clinic, Barcelona). Informed written consent was obtained from all individuals.

Anthropometrical and functional assessment

Anthropometrical, lung and muscle function evaluations were conducted as previously described [27,28].

Muscle biopsies and blood samples

Vastus lateralis biopsies

Muscle samples were obtained from the VL of all groups of patients and control subjects using the open muscle biopsy technique, as described previously [28].

Molecular biology analyses

RNA isolation

Total RNA was first isolated from snap-frozen skeletal muscles using Trizol reagent following the manufacturer's protocol (Life technologies). miRNA and mRNA reverse transcription (RT): miRNA RT was performed using TaqMan® miRNA assays (Life Technologies) following the manufacturer's instructions. Real time-PCR amplification (qRT–PCR): TaqMan based qPCR reactions were performed using the ABI PRISM 7900HT Sequence Detector System (Applied BioSystems) together with a commercially available predesigned miRNA assay, primers and probes as shown in Table 1 [29]. Results in the COPD groups are shown as the relative expression of that in the control group, which was set to equal to 1.

Table 1
miRNA assays and probes used for the quantitative analyses of the target genes using real-time PCR

Abbreviations: EP300, E1A-binding protein p300; FST, follistatin; hsa, homo sapiens; ID, identification; miR, miRNA; MKL, megakaryoblastic leukaemia; MSTN, myostatin; MyoD, myogenic differentiation 1; MYOG, myogenin; snRNA, small nuclear RNA.

Assay NameAssay IDmiRBase accession number
Muscle-specific, myomiRs    
hsamiR-1 002222 MIMAT0000416  
hsamiR-133a 002246 MIMAT0000427  
hsamiR-206 000510 MIMAT0000462  
Other miRNAs (muscle-enriched)    
hsamiR-486 001278 MIMAT0002177  
hsamiR-27a 000408 MIMAT0000084  
hsamiR-29b 000413 MIMAT0000100  
hsamiR-181a 000480 MIMAT0000256  
  NCBI Accession number  
U6 snRNA, housekeeping gene 001973 NR_004394  
Gene SymbolAssay IDTaqMan probe context sequence (5′-3′)Genbank accession number
EP300 Hs00914223_m1 CACCATGGAGAAGCATAAAGAGGTC NM_001429.3 
MKL1 Hs00252979_m1 CCATCATTGTGGGCCAGGTGAACTA NM_020831.3 
MKL2 Hs00393603_m1 TGCCTGCACCATTCAAGCCACTCAA NM_014048.3 
SRF Hs00182371_m1 CATGAAGAAGGCCTATGAGCTGTCC NM_003131.2 
IGF-1 Hs00153126_m1 TTGTGATTTCTTGAAGGTGAAGATG NM_000618.3 
AKT Hs00178289_m1 CTCCTGAGGAGCGGGAGGAGTGGAC NM_001014431.1 
FST Hs00246256_m1 GCAGTGCCCAGGCTGGGAACTGCTG NM_006350.3 
MSTN Hs00976237_m1 ATGCCTACAGAGTCTGATTTTCTAA NM_005259.2 
MyoD Hs00159528_m1 GGCGCCCAGCGAACCCAGGCCCGGG NM_002478.4 
MYOG Hs01072232_m1 CCAACCCAGGGGATCATCTGCTCAC NM_002479.5 
PAX3 Hs00240950_m1 GCAAGCCCAAGCAGGTGACAACGCC NM_000438.5 
HOXA11 Hs00194149_m1 GGCCGGCGGCTCCAGTGGCCAACGC NM_005523.5 
SUMO2 Hs02743873_g1 CATTGTAAAACCAAGGACAATTTTA NM_006937.3 
SUMO3 Hs00739248_m1 GCGAGAGGCAGGGCTTGTCAATGAG NM_006936.2 
GAPDH Hs99999905_m1 GGGCGCCTGGTCACCAGGGCTGCTT NM_002046.4 
Assay NameAssay IDmiRBase accession number
Muscle-specific, myomiRs    
hsamiR-1 002222 MIMAT0000416  
hsamiR-133a 002246 MIMAT0000427  
hsamiR-206 000510 MIMAT0000462  
Other miRNAs (muscle-enriched)    
hsamiR-486 001278 MIMAT0002177  
hsamiR-27a 000408 MIMAT0000084  
hsamiR-29b 000413 MIMAT0000100  
hsamiR-181a 000480 MIMAT0000256  
  NCBI Accession number  
U6 snRNA, housekeeping gene 001973 NR_004394  
Gene SymbolAssay IDTaqMan probe context sequence (5′-3′)Genbank accession number
EP300 Hs00914223_m1 CACCATGGAGAAGCATAAAGAGGTC NM_001429.3 
MKL1 Hs00252979_m1 CCATCATTGTGGGCCAGGTGAACTA NM_020831.3 
MKL2 Hs00393603_m1 TGCCTGCACCATTCAAGCCACTCAA NM_014048.3 
SRF Hs00182371_m1 CATGAAGAAGGCCTATGAGCTGTCC NM_003131.2 
IGF-1 Hs00153126_m1 TTGTGATTTCTTGAAGGTGAAGATG NM_000618.3 
AKT Hs00178289_m1 CTCCTGAGGAGCGGGAGGAGTGGAC NM_001014431.1 
FST Hs00246256_m1 GCAGTGCCCAGGCTGGGAACTGCTG NM_006350.3 
MSTN Hs00976237_m1 ATGCCTACAGAGTCTGATTTTCTAA NM_005259.2 
MyoD Hs00159528_m1 GGCGCCCAGCGAACCCAGGCCCGGG NM_002478.4 
MYOG Hs01072232_m1 CCAACCCAGGGGATCATCTGCTCAC NM_002479.5 
PAX3 Hs00240950_m1 GCAAGCCCAAGCAGGTGACAACGCC NM_000438.5 
HOXA11 Hs00194149_m1 GGCCGGCGGCTCCAGTGGCCAACGC NM_005523.5 
SUMO2 Hs02743873_g1 CATTGTAAAACCAAGGACAATTTTA NM_006937.3 
SUMO3 Hs00739248_m1 GCGAGAGGCAGGGCTTGTCAATGAG NM_006936.2 
GAPDH Hs99999905_m1 GGGCGCCTGGTCACCAGGGCTGCTT NM_002046.4 

Histone extraction

Total histone content was extracted from snap-frozen skeletal muscles using the EpiQuik™ Total Histone Extraction Kit (Epigentek) following the precise manufacturer's instructions for solid tissues and as previously reported [30].

Histone H3 and H4 acetylation assays

Total acetylation levels of histones H3 and H4 were quantified separately in the VL of both patients and healthy controls using the EpiQuik™ Total Histone H3 and H4 Acetylation Detection Fast Kits (Epigentek) respectively, following the precise manufacturer's instructions and previous studies [31,32]. In the samples, the minimum detectable concentration of histones H3 and H4 acetylated-lysine sites was set to be 5 ng/well of histone extract. Data are expressed as the total amount of acetylated H3 (ng/mg protein) and acetylated H4 (ng/mg protein) separately.

Immunoblotting of 1D electrophoresis

Protein levels of the different molecular markers analysed in the muscles of all the study subjects were explored by means of immunoblotting procedures as previously described [28].

Muscle fibre counts and morphometry

On 3-μm muscle paraffin-embedded sections from VL muscles of all study groups, MyHC-I and–II isoforms were identified using specific antibodies as published elsewhere [28].

Statistical analysis

Statistical power was calculated using specific software (StudySize 2.0, CreoStat HB). QMVC (quadriceps maximal velocity contraction) was selected as the target variable on the basis of the ANOVA test to estimate the statistical power in the study. On the basis of a standard power statistics established at a minimum of 80% and assuming an α-error of 0.05, the statistical power was sufficiently high to detect a minimum difference of two points between groups in the sample size and S.D.

Normality of the study variables was checked using the Shapiro–Wilk test. In order to fulfill the study hypothesis two different types of comparisons and statistical tests were used: (1) comparisons between all COPD patients as a group and healthy subjects were assessed using the Student's t test and (2) comparisons between the two groups of COPD patients, non-muscle weakness and muscle-weakness and healthy subjects were analysed using one-way ANOVA, in which Tukey's post-hoc analysis was used to adjust for multiple comparisons. Pearson's Chi-Square test was employed to assess the statistical significant differences in smoking history variables between the two groups of patients and healthy controls. Correlations between clinical, physiological and biological variables were explored using the Pearson's correlation coefficient among all COPD patients as a group and in muscle-weakness patients separately. Data are presented as mean and S.D. in the tables, whereas they are shown as box and whisker plots in the figures. A level of significance of P≤0.05 was established. Statistical analyses were performed using the Statistical Package for the Social Sciences (Portable SPSS, PASW Statistics 18.0 version for windows, SPSS Inc).

RESULTS

Clinical characteristics

Table 2 illustrates all clinical and functional variables of controls and COPD patients recruited in the study. Age did not significantly differ among the study subjects. Body composition as measured by body mass index (BMI) and fat-free mass index (FFMI) was significantly reduced in all COPD patients together. Moreover, these two parameters and body weight were significantly reduced in muscle-weakness COPD patients compared with both healthy controls and non-muscle-weakness patients, whereas no differences were observed between non-muscle-weakness patients and the controls. Body weight change, expressed as kilograms lost in the previous year was also significantly reduced in muscle-weakness COPD patients compared with both non-muscle-weakness patients and healthy controls. Importantly, smoking history did not differ between any group of COPD patients and the controls. However, the proportions of active smokers were significantly reduced in muscle-weakness COPD patients compared with non-muscle-weakness patients (39% reduction). Muscle-weakness patients exhibited very severe airflow limitation and functional signs of emphysema, whereas non-muscle-weakness patients showed a moderate-to-severe airflow obstruction. Compared with healthy controls, in both groups of COPD patients, exercise capacity and quadriceps strength were significantly decreased, which were especially impaired in the muscle-weakness group. Among all COPD patients together, significant correlations were found between QMVC and walking test distance, peak work rate and oxygen uptake (r=0.361, P=0.022; r=0.691, P=0.003; and r=0.841, P<0.001 respectively). Levels of C-reactive protein, fibrinogen and globular sedimentation velocity were higher in all COPD patients than in healthy controls, particularly in the muscle-weakness patients.

Table 2
Anthropometric characteristics and functional status of the study subjects

Values are expressed as mean (S.D.).

Abbreviations: CRP, C-reactive protein; DLco, carbon monoxide transfer; FEV1, forced expiratory volume in 1 s; FVC, forced vital capacity; GSV, globular sedimentation velocity; KCO, Krough transfer factor; N, number of patients; PaCO2, arterial carbon dioxide partial pressure; PaO2, arterial oxygen partial pressure; pred, predicted; RV, residual volume; TLC, total lung capacity; VO2 peak, peak exercise oxygen uptake; WR peak, peak work rate.

Statistical analyses: Comparisons between all patients with COPD and the control subjects were assessed using the Student's ttest; comparisons among the two groups of COPD patients, non-muscle-weakness and the muscle-weakness and healthy subjects were analysed using one-way ANOVA and Tukey's post-hoc analysis to adjust for multiple comparisons.

Statistical significance: *P≤0.05, **P≤0.01, ***P≤0.001 between all patients with COPD and the control subjects; §P≤0.05, §§P≤0.01, §§§P≤0.001 between either the non-muscle-weakness or the muscle-weakness COPD patients and the healthy controls; P≤0.05, ¶¶P≤0.01, ¶¶¶P≤0.001 between the muscle-weakness COPD patients and the non-muscle-weakness patients.

COPD patients
Controls N=19All COPD patients N=41Non-muscle weakness N=16Muscle weakness N=25
Anthropometry     
Age (years) 65 (8) 68 (6) 68 (6) 68 (5) 
BMI (kg/m226 (3) 24 (5)* 26 (4) 22 (4)§§¶¶ 
FFMI (kg/m219 (2) 16 (2)*** 18 (2) 16 (2)§§§¶ 
Body weight (kg) 73 (8) 69 (13) 76 (12) 66 (13)§§§ 
Body weight change (kg/year) 0 (0) −1.8 (2.4)*** −0.3 (2.5) −2.8 (1.7)§§§¶¶¶ 
Smoking history     
Active, N (%) 6, 32 16, 39 10, 63 6, 24 
Ex-smoker, N (%) 8, 42 25, 61 6, 37 19, 76 
Never smoker, N (%) 5, 26 0, 0 0, 0 0, 0 
Packs-year 54 (20) 61 (24) 59 (28) 62 (22) 
Lung function     
FEV1 (% pred) 93 (12) 34 (15)*** 51 (6)§§§ 23 (5)§§§¶¶¶ 
FVC (% pred) 88 (9) 58 (18)*** 76 (16) 50 (12)§§§¶¶¶ 
FEV1/FVC (%) 73 (4) 44 (11)*** 53 (9)§§§ 38 (9)§§§¶¶¶ 
RV (% pred) 105 (18) 198 (69)*** 139 (44) 220 (63)§§§¶¶ 
TLC (% pred) 101 (12) 109 (16) 105 (18) 111 (15) 
RV/TLC 49 (23) 64 (11)*** 54 (7) 70 (7)§§§¶¶ 
DLco (% pred) 89 (14) 52 (26)*** 71 (21)§ 38 (21)§§§¶¶¶ 
KCO (% pred) 87 (16) 63 (20)*** 71 (19) 59 (20)§§§ 
PaO2 (kPa) 11.6 (1.1) 9.2 (1.2)*** 9.5 (1.3)§§ 9.0 (1)§§§ 
PaCO2 (kPa) 5.2 (0.5) 5.6 (0.7)* 5.4 (0.8) 5.8 (0.5) 
Exercise capacity and muscle function     
VO2 peak (% pred) 87 (10) 48 (23)*** 67 (16)§ 31 (10)§§§¶¶¶ 
WR peak (% pred) 81 (21) 48 (26)*** 64 (26)§ 31 (12)§§§¶ 
6-min walking test (m) 508 (71) 412 (92)*** 459 (69) 380 (93)§§§¶ 
QMVC (kg) 39 (2) 29 (3)*** 32 (1)§§§ 27 (1)§§§¶¶¶ 
Blood parameters     
Albumin (g/dl) 4.3 (0.4) 4.4 (0.4) 4.3 (0.3) 4.4 (0.5) 
Total proteins (g/dl) 7.2 (0.5) 7.3 (0.5) 7.1 (0.6) 7.3 (0.5) 
CRP (mg/dl) 0.3 (0.2) 1.5 (2.2)** 0.5 (0.4) 2.5 (2.7)§§¶¶ 
Fibrinogen (mg/dl) 311 (35) 412 (75)*** 363 (53) 434 (74)§§§¶ 
GSV (mm/h) 6 (4) 28 (19)*** 20 (11)§ 32 (22)§§§¶ 
COPD patients
Controls N=19All COPD patients N=41Non-muscle weakness N=16Muscle weakness N=25
Anthropometry     
Age (years) 65 (8) 68 (6) 68 (6) 68 (5) 
BMI (kg/m226 (3) 24 (5)* 26 (4) 22 (4)§§¶¶ 
FFMI (kg/m219 (2) 16 (2)*** 18 (2) 16 (2)§§§¶ 
Body weight (kg) 73 (8) 69 (13) 76 (12) 66 (13)§§§ 
Body weight change (kg/year) 0 (0) −1.8 (2.4)*** −0.3 (2.5) −2.8 (1.7)§§§¶¶¶ 
Smoking history     
Active, N (%) 6, 32 16, 39 10, 63 6, 24 
Ex-smoker, N (%) 8, 42 25, 61 6, 37 19, 76 
Never smoker, N (%) 5, 26 0, 0 0, 0 0, 0 
Packs-year 54 (20) 61 (24) 59 (28) 62 (22) 
Lung function     
FEV1 (% pred) 93 (12) 34 (15)*** 51 (6)§§§ 23 (5)§§§¶¶¶ 
FVC (% pred) 88 (9) 58 (18)*** 76 (16) 50 (12)§§§¶¶¶ 
FEV1/FVC (%) 73 (4) 44 (11)*** 53 (9)§§§ 38 (9)§§§¶¶¶ 
RV (% pred) 105 (18) 198 (69)*** 139 (44) 220 (63)§§§¶¶ 
TLC (% pred) 101 (12) 109 (16) 105 (18) 111 (15) 
RV/TLC 49 (23) 64 (11)*** 54 (7) 70 (7)§§§¶¶ 
DLco (% pred) 89 (14) 52 (26)*** 71 (21)§ 38 (21)§§§¶¶¶ 
KCO (% pred) 87 (16) 63 (20)*** 71 (19) 59 (20)§§§ 
PaO2 (kPa) 11.6 (1.1) 9.2 (1.2)*** 9.5 (1.3)§§ 9.0 (1)§§§ 
PaCO2 (kPa) 5.2 (0.5) 5.6 (0.7)* 5.4 (0.8) 5.8 (0.5) 
Exercise capacity and muscle function     
VO2 peak (% pred) 87 (10) 48 (23)*** 67 (16)§ 31 (10)§§§¶¶¶ 
WR peak (% pred) 81 (21) 48 (26)*** 64 (26)§ 31 (12)§§§¶ 
6-min walking test (m) 508 (71) 412 (92)*** 459 (69) 380 (93)§§§¶ 
QMVC (kg) 39 (2) 29 (3)*** 32 (1)§§§ 27 (1)§§§¶¶¶ 
Blood parameters     
Albumin (g/dl) 4.3 (0.4) 4.4 (0.4) 4.3 (0.3) 4.4 (0.5) 
Total proteins (g/dl) 7.2 (0.5) 7.3 (0.5) 7.1 (0.6) 7.3 (0.5) 
CRP (mg/dl) 0.3 (0.2) 1.5 (2.2)** 0.5 (0.4) 2.5 (2.7)§§¶¶ 
Fibrinogen (mg/dl) 311 (35) 412 (75)*** 363 (53) 434 (74)§§§¶ 
GSV (mm/h) 6 (4) 28 (19)*** 20 (11)§ 32 (22)§§§¶ 

Biological markers in muscles

miRNAs

Compared with healthy controls, levels of miR-1 and miR-206 were significantly up-regulated in the VL of all COPD patients and in muscle-weakness patients, miR-206 expression was also significantly greater than in non-muscle-weakness patients (Figures 1A and 1B). Nonetheless, the expression of miR-133 did not differ in muscles of the study groups (Figure 1C). Compared with healthy controls, expression levels of miR-27 were significantly higher in VL of all COPD patients and in muscle-weakness patients (Figure 1D). Expression levels of miR-486, miR-29b and miR-181a did not significantly differ in limb muscles between any of the study groups (Figures 1E–1G respectively). Importantly, a significant negative correlation was detected between QMVC and levels of miR-206 (r=−0.433, P=0.006) when all COPD patients were analysed together.

Expression of muscle-enriched miRNAs in the VL of patients with COPD with and without muscle weakness and healthy controls
Figure 1
Expression of muscle-enriched miRNAs in the VL of patients with COPD with and without muscle weakness and healthy controls

(A) Standard box plots with median (twenty-fifth and seventy-fifth percentiles) and whiskers (at minimum and maximum values) of miR-1 expression are depicted in the VL of COPD patients and control subjects, outliers are represented by open dots. Statistical significance is represented as follows: **P≤0.01 between all the patients with COPD and the control subjects; n.s., non-significant between the non-muscle-weakness COPD patients and the healthy controls; §P≤0.05 between the muscle-weakness COPD patients and the healthy subjects. Outliers are represented as open dots. (B) Standard box plots with median (twenty-fifth and seventy-fifth percentiles) and whiskers (at minimum and maximum values) of miR-206 expression are depicted in the VL of the COPD patients and the control subjects, outliers are represented by open dots. Statistical significance is represented as follows: *P<0.05 between all patients with COPD and the control subjects; n.s., non-significant between the non-muscle-weakness COPD patients and the healthy controls; §§P<0.01 between the muscle-weakness COPD patients and the healthy subjects; ¶P<0.05 between the muscle-weakness COPD patients and the non-muscle-weakness patients. Outliers are represented as open dots. (C) Standard box plots with median (twenty-fifth and seventy-fifth percentiles) and whiskers (at minimum and maximum values) of miR-133 expression are depicted in the VL of COPD patients and control subjects, outliers are represented by open dots. Statistical significance is represented as follows: n.s., non-significant among any of the study groups. Outliers are represented as open dots. (D) Standard box plots with median (twenty-fifth and seventy-fifth percentiles) and whiskers (at minimum and maximum values) of miR-27a expression are depicted in the VL of the COPD patients and the control subjects. Statistical significance is represented as follows: **P<0.01 between all patients with COPD and the control subjects; n.s., non-significant between non-muscle weakness COPD patients and the healthy controls; §P<0.05 between the muscle-weakness COPD patients and the healthy subjects. (E) Standard box plots with median (twenty-fifth and seventy-fifth percentiles) and whiskers (at minimum and maximum values) of miR-486 expression are depicted in the VL of the COPD patients and the control subjects, outliers are represented by open dots. Statistical significance is represented as follows: n.s., non-significant among any of the study groups. Outliers are represented as open dots. (F) Standard box plots with median (twenty-fifth and seventy-fifth percentiles) and whiskers (at minimum and maximum values) of miR-29b expression are depicted in the VL of the COPD patients and the control subjects. Statistical significance is represented as follows: n.s., non-significant among any of the study groups. (G) Standard box plots with median (twenty-fifth and seventy-fifth percentiles) and whiskers (at minimum and maximum values) of miR-181a expression are depicted in the VL of the COPD patients and the control subjects. Statistical significance is represented as follows: n.s., non-significant among any of the study groups.

Figure 1
Expression of muscle-enriched miRNAs in the VL of patients with COPD with and without muscle weakness and healthy controls

(A) Standard box plots with median (twenty-fifth and seventy-fifth percentiles) and whiskers (at minimum and maximum values) of miR-1 expression are depicted in the VL of COPD patients and control subjects, outliers are represented by open dots. Statistical significance is represented as follows: **P≤0.01 between all the patients with COPD and the control subjects; n.s., non-significant between the non-muscle-weakness COPD patients and the healthy controls; §P≤0.05 between the muscle-weakness COPD patients and the healthy subjects. Outliers are represented as open dots. (B) Standard box plots with median (twenty-fifth and seventy-fifth percentiles) and whiskers (at minimum and maximum values) of miR-206 expression are depicted in the VL of the COPD patients and the control subjects, outliers are represented by open dots. Statistical significance is represented as follows: *P<0.05 between all patients with COPD and the control subjects; n.s., non-significant between the non-muscle-weakness COPD patients and the healthy controls; §§P<0.01 between the muscle-weakness COPD patients and the healthy subjects; ¶P<0.05 between the muscle-weakness COPD patients and the non-muscle-weakness patients. Outliers are represented as open dots. (C) Standard box plots with median (twenty-fifth and seventy-fifth percentiles) and whiskers (at minimum and maximum values) of miR-133 expression are depicted in the VL of COPD patients and control subjects, outliers are represented by open dots. Statistical significance is represented as follows: n.s., non-significant among any of the study groups. Outliers are represented as open dots. (D) Standard box plots with median (twenty-fifth and seventy-fifth percentiles) and whiskers (at minimum and maximum values) of miR-27a expression are depicted in the VL of the COPD patients and the control subjects. Statistical significance is represented as follows: **P<0.01 between all patients with COPD and the control subjects; n.s., non-significant between non-muscle weakness COPD patients and the healthy controls; §P<0.05 between the muscle-weakness COPD patients and the healthy subjects. (E) Standard box plots with median (twenty-fifth and seventy-fifth percentiles) and whiskers (at minimum and maximum values) of miR-486 expression are depicted in the VL of the COPD patients and the control subjects, outliers are represented by open dots. Statistical significance is represented as follows: n.s., non-significant among any of the study groups. Outliers are represented as open dots. (F) Standard box plots with median (twenty-fifth and seventy-fifth percentiles) and whiskers (at minimum and maximum values) of miR-29b expression are depicted in the VL of the COPD patients and the control subjects. Statistical significance is represented as follows: n.s., non-significant among any of the study groups. (G) Standard box plots with median (twenty-fifth and seventy-fifth percentiles) and whiskers (at minimum and maximum values) of miR-181a expression are depicted in the VL of the COPD patients and the control subjects. Statistical significance is represented as follows: n.s., non-significant among any of the study groups.

Histone modifications

Compared with healthy controls, muscle levels of both total and histone lysine-acetylated proteins were significantly increased in the VL of all COPD patients together and in muscle-weakness patients (Figures 2A and 2B respectively). Additionally, muscle levels of total lysine-acetylated proteins were also higher in muscle-weakness patients compared with non-muscle weakness patients (Figure 2A). Expression levels of HAT p300 did not differ between any of the study groups (Figure 2C). Interestingly, when all COPD patients were analysed as a group, significant negative correlations were found between QMVC and levels of total lysine-acetylated proteins (r=−0.442, P=0.021) and lysine-acetylated histones (r=−0.455, P=0.033), which also positively correlated between them (r=0.577, P=0.024). Total lysine-acetylated protein levels also significantly correlated with BMI among all COPD patients (r=−0.496, P=0.009). Compared with controls, expression levels of acetylated histone 3 were significantly greater in muscles of all COPD patients and in muscle-weakness patients, whereas no differences were observed in acetylated histone 4 among the study groups (Figures 2D and 2E respectively). A significant negative correlation was detected between acetylated histone 3 levels and FFMI among all COPD patients together (r=−0.560, P=0.047).

Expression of total protein and histone acetylation in the VL of patients with COPD with and without muscle weakness and healthy controls
Figure 2
Expression of total protein and histone acetylation in the VL of patients with COPD with and without muscle weakness and healthy controls

(A) Standard box plots with median (twenty-fifth and seventy-fifth percentiles) and whiskers (at minimum and maximum values) of total lysine-acetylated proteins are depicted in the VL of the COPD patients and control subjects, outliers are represented by open dots. Statistical significance is represented as follows: **P<0.01 between all patients with COPD and the control subjects; n.s., non-significant between the non-muscle-weakness COPD patients and the healthy controls; §§§P<0.001 between the muscle-weakness COPD patients and the healthy subjects; ¶P<0.05 between the muscle-weakness COPD patients and the non-muscle weakness patients. Outliers are represented as open dots. (B) Standard box plots with median (twenty-fifth and seventy-fifth percentiles) and whiskers (at minimum and maximum values) of lysine-acetylated histones are depicted in the VL of COPD patients and control subjects, outliers are represented by open dots. Statistical significance is represented as follows: *P<0.05 between all patients with COPD and the control subjects; n.s., non-significant between non-muscle weakness COPD patients and the healthy controls; §P<0.05 between muscle-weakness COPD patients and the healthy subjects. Outliers are represented as open dots. (C) Standard box plots with median (twenty-fifth and seventy-fifth percentiles) and whiskers (at minimum and maximum values) of the nuclear cofactor p300 expression are depicted in the VL of the COPD patients and the control subjects, outliers are represented by open dots. Statistical significance is represented as follows: n.s., non-significant among any of the study groups. Outliers are represented as open dots. (D) Standard box plots with median (twenty-fifth and seventy-fifth percentiles) and whiskers (at minimum and maximum values) of total amount of lysine-acetylated histone 3 are depicted in the VL of the COPD patients and the control subjects. Statistical significance is represented as follows: *P<0.05 between all patients with COPD and the control subjects; n.s., non-significant between the non-muscle-weakness COPD patients and the healthy controls; §P<0.05 between muscle-weakness the COPD patients and the healthy subjects. (E) Standard box plots with median (twenty-fifth and seventy-fifth percentiles) and whiskers (at minimum and maximum values) of total amount of lysine-acetylated histone 4 are depicted in the VL of the COPD patients and the control subjects, outliers are represented by open dots. Statistical significance is represented as follows: n.s., non-significant among any of the study groups. Outliers are represented as open dots.

Figure 2
Expression of total protein and histone acetylation in the VL of patients with COPD with and without muscle weakness and healthy controls

(A) Standard box plots with median (twenty-fifth and seventy-fifth percentiles) and whiskers (at minimum and maximum values) of total lysine-acetylated proteins are depicted in the VL of the COPD patients and control subjects, outliers are represented by open dots. Statistical significance is represented as follows: **P<0.01 between all patients with COPD and the control subjects; n.s., non-significant between the non-muscle-weakness COPD patients and the healthy controls; §§§P<0.001 between the muscle-weakness COPD patients and the healthy subjects; ¶P<0.05 between the muscle-weakness COPD patients and the non-muscle weakness patients. Outliers are represented as open dots. (B) Standard box plots with median (twenty-fifth and seventy-fifth percentiles) and whiskers (at minimum and maximum values) of lysine-acetylated histones are depicted in the VL of COPD patients and control subjects, outliers are represented by open dots. Statistical significance is represented as follows: *P<0.05 between all patients with COPD and the control subjects; n.s., non-significant between non-muscle weakness COPD patients and the healthy controls; §P<0.05 between muscle-weakness COPD patients and the healthy subjects. Outliers are represented as open dots. (C) Standard box plots with median (twenty-fifth and seventy-fifth percentiles) and whiskers (at minimum and maximum values) of the nuclear cofactor p300 expression are depicted in the VL of the COPD patients and the control subjects, outliers are represented by open dots. Statistical significance is represented as follows: n.s., non-significant among any of the study groups. Outliers are represented as open dots. (D) Standard box plots with median (twenty-fifth and seventy-fifth percentiles) and whiskers (at minimum and maximum values) of total amount of lysine-acetylated histone 3 are depicted in the VL of the COPD patients and the control subjects. Statistical significance is represented as follows: *P<0.05 between all patients with COPD and the control subjects; n.s., non-significant between the non-muscle-weakness COPD patients and the healthy controls; §P<0.05 between muscle-weakness the COPD patients and the healthy subjects. (E) Standard box plots with median (twenty-fifth and seventy-fifth percentiles) and whiskers (at minimum and maximum values) of total amount of lysine-acetylated histone 4 are depicted in the VL of the COPD patients and the control subjects, outliers are represented by open dots. Statistical significance is represented as follows: n.s., non-significant among any of the study groups. Outliers are represented as open dots.

Importantly, compared with healthy controls, levels of HDAC3 and HDAC4 were significantly decreased in VL of all COPD patients together and in muscle-weakness patients, but not in the non-muscle-weakness group (Figures 3A and 3B respectively). Levels of HDAC3 positively correlated with FFMI (r=0.566, P=0.028). Protein levels of SIRT-1 (sirtuin-1) were significantly reduced only in limb muscles of muscle-weakness patients compared with healthy controls (Figure 3C). Levels of HDAC6 did not differ in muscles among the study groups (Figure 3D). Moreover, among all COPD patients, significant correlations were observed between muscle levels of SIRT-1 and total acetylated proteins and FFMI (r=−0.441, P=0.024 and r=0.558, P=0.031 respectively).

Expression of HDACs in the VL of the patients with COPD with and without muscle weakness and healthy controls
Figure 3
Expression of HDACs in the VL of the patients with COPD with and without muscle weakness and healthy controls

(A) Standard box plots with median (twenty-fifth and seventy-fifth percentiles) and whiskers (at minimum and maximum values) of HDAC3 content are depicted in the VL of the COPD patients and control subjects, outliers are represented by open dots. Statistical significance is represented as follows: **P<0.01 between all patients with COPD and the control subjects; n.s., non-significant between the non-muscle-weakness COPD patients and the healthy controls; §§P<0.01 between the muscle-weakness COPD patients and the healthy subjects. Outliers are represented as open dots. (B) Standard box plots with median (twenty-fifth and seventy-fifth percentiles) and whiskers (at minimum and maximum values) of HDAC4 content are depicted in the VL of the COPD patients and the control subjects. Statistical significance is represented as follows: *P<0.05 between all patients with COPD and the control subjects; n.s., non-significant between the non-muscle-weakness COPD patients and the healthy controls; §P<0.05 between the muscle-weakness COPD patients and the healthy subjects. (C) Standard box plots with median (twenty-fifth and seventy-fifth percentiles) and whiskers (at minimum and maximum values) of SIRT-1 protein levels are depicted in the VL of the COPD patients and the control subjects. Statistical significance is represented as follows: n.s., non-significant between both all patients with COPD and non-muscle-weakness COPD patients and the healthy controls; §P<0.05 between the muscle-weakness COPD patients and the healthy subjects. (D) Standard box plots with median (twenty-fifth and seventy-fifth percentiles) and whiskers (at minimum and maximum values) of HDAC6 content are depicted in the VL of the COPD patients and the control subjects, outliers are represented by open dots. Statistical significance is represented as follows: n.s., non-significant among any of the study groups. Outliers are represented as open dots.

Figure 3
Expression of HDACs in the VL of the patients with COPD with and without muscle weakness and healthy controls

(A) Standard box plots with median (twenty-fifth and seventy-fifth percentiles) and whiskers (at minimum and maximum values) of HDAC3 content are depicted in the VL of the COPD patients and control subjects, outliers are represented by open dots. Statistical significance is represented as follows: **P<0.01 between all patients with COPD and the control subjects; n.s., non-significant between the non-muscle-weakness COPD patients and the healthy controls; §§P<0.01 between the muscle-weakness COPD patients and the healthy subjects. Outliers are represented as open dots. (B) Standard box plots with median (twenty-fifth and seventy-fifth percentiles) and whiskers (at minimum and maximum values) of HDAC4 content are depicted in the VL of the COPD patients and the control subjects. Statistical significance is represented as follows: *P<0.05 between all patients with COPD and the control subjects; n.s., non-significant between the non-muscle-weakness COPD patients and the healthy controls; §P<0.05 between the muscle-weakness COPD patients and the healthy subjects. (C) Standard box plots with median (twenty-fifth and seventy-fifth percentiles) and whiskers (at minimum and maximum values) of SIRT-1 protein levels are depicted in the VL of the COPD patients and the control subjects. Statistical significance is represented as follows: n.s., non-significant between both all patients with COPD and non-muscle-weakness COPD patients and the healthy controls; §P<0.05 between the muscle-weakness COPD patients and the healthy subjects. (D) Standard box plots with median (twenty-fifth and seventy-fifth percentiles) and whiskers (at minimum and maximum values) of HDAC6 content are depicted in the VL of the COPD patients and the control subjects, outliers are represented by open dots. Statistical significance is represented as follows: n.s., non-significant among any of the study groups. Outliers are represented as open dots.

Muscle growth pathway

Compared with healthy controls, expression levels of insulin-like growth factor (IGF)-1 were significantly reduced in VL of all COPD patients and in muscle-weakness patients, whereas levels of Akt (v-akt murine thymoma viral oncogene homologue 1) did not differ between any of the study groups (Figures 4A and 4B respectively). Interestingly, compared with healthy controls, expression levels of follistatin were significantly greater in muscles of all COPD patients together and in muscle-weakness patients, whereas muscle myostatin expression was reciprocally reduced in the same patient groups (Figures 4C and 4D respectively). Interestingly, among all COPD patients, significant correlations were found between QMVC and muscle follistatin and myostatin expression levels (r=−0.406, P=0.026 and r=0.343, P=0.043 respectively). In muscle-weakness patients, muscle follistatin expression levels also correlated with QMVC (r=−0.700, P=0.002).

Expression of the muscle growth markers in the VL of the patients with COPD with and without muscle weakness and the healthy controls
Figure 4
Expression of the muscle growth markers in the VL of the patients with COPD with and without muscle weakness and the healthy controls

(A) Standard box plots with median (twenty-fifth and seventy-fifth percentiles) and whiskers (at minimum and maximum values) of IGF-1 expression are depicted in the VL of the COPD patients and the control subjects, outliers are represented by open dots. Statistical significance is represented as follows: *P<0.05 between all patients with COPD and the control subjects; n.s., non-significant between non-muscle weakness COPD patients and the healthy controls; §P<0.05 between the muscle-weakness COPD patients and the healthy subjects. Outliers are represented as open dots. (B) Standard box plots with median (twenty-fifth and seventy-fifth percentiles) and whiskers (at minimum and maximum values) of Akt expression are depicted in the VL of the COPD patients and the control subjects, outliers are represented by open dots. Statistical significance is represented as follows: n.s., non-significant among any of the study groups. Outliers are represented as open dots. (C) Standard box plots with median (twenty-fifth and seventy-fifth percentiles) and whiskers (at minimum and maximum values) of follistatin expression are depicted in the VL of the COPD patients and the control subjects. Statistical significance is represented as follows: **P<0.01 between all patients with COPD and the control subjects; n.s., non-significant between non-muscle-weakness COPD patients and the healthy controls; §§P<0.01 between muscle-weakness COPD patients and the healthy subjects. (D) Standard box plots with median (twenty-fifth and seventy-fifth percentiles) and whiskers (at minimum and maximum values) of myostatin expression are depicted in the VL of the COPD patients and the control subjects. Statistical significance is represented as follows: **P<0.01 between all patients with COPD and the control subjects; n.s., non-significant between the non-muscle weakness COPD patients and the healthy controls; §§§P<0.001 between the muscle-weakness COPD patients and the healthy subjects.

Figure 4
Expression of the muscle growth markers in the VL of the patients with COPD with and without muscle weakness and the healthy controls

(A) Standard box plots with median (twenty-fifth and seventy-fifth percentiles) and whiskers (at minimum and maximum values) of IGF-1 expression are depicted in the VL of the COPD patients and the control subjects, outliers are represented by open dots. Statistical significance is represented as follows: *P<0.05 between all patients with COPD and the control subjects; n.s., non-significant between non-muscle weakness COPD patients and the healthy controls; §P<0.05 between the muscle-weakness COPD patients and the healthy subjects. Outliers are represented as open dots. (B) Standard box plots with median (twenty-fifth and seventy-fifth percentiles) and whiskers (at minimum and maximum values) of Akt expression are depicted in the VL of the COPD patients and the control subjects, outliers are represented by open dots. Statistical significance is represented as follows: n.s., non-significant among any of the study groups. Outliers are represented as open dots. (C) Standard box plots with median (twenty-fifth and seventy-fifth percentiles) and whiskers (at minimum and maximum values) of follistatin expression are depicted in the VL of the COPD patients and the control subjects. Statistical significance is represented as follows: **P<0.01 between all patients with COPD and the control subjects; n.s., non-significant between non-muscle-weakness COPD patients and the healthy controls; §§P<0.01 between muscle-weakness COPD patients and the healthy subjects. (D) Standard box plots with median (twenty-fifth and seventy-fifth percentiles) and whiskers (at minimum and maximum values) of myostatin expression are depicted in the VL of the COPD patients and the control subjects. Statistical significance is represented as follows: **P<0.01 between all patients with COPD and the control subjects; n.s., non-significant between the non-muscle weakness COPD patients and the healthy controls; §§§P<0.001 between the muscle-weakness COPD patients and the healthy subjects.

Myogenic transcription factors

Muscle expression levels of MRTF-A and MRTF-B did not differ between patients and controls (Figures 5A and 5B respectively), whereas expression levels of SRF were up-regulated in VL of all COPD patients compared with control subjects (Figure 5C). Muscle protein levels of myocyte enhancer factors (MEF)2C, MEF2D, myoD, myogenin, homoeobox protein (Hox)A11 did not differ between the study groups (Table 3). Compared with healthy controls, levels of YY1 were greater in VL of muscle-weakness COPD patients, whereas paired box (Pax)3 levels were decreased in all COPD patients as a group and in muscle-weakness patients. A significant correlation was found between muscle YY1 levels and QMVC among all COPD patients (r=−0.434, P=0.027).

Expression of myogenic transcription factors in the VL of patients with COPD with and without muscle weakness and healthy controls
Figure 5
Expression of myogenic transcription factors in the VL of patients with COPD with and without muscle weakness and healthy controls

(A) Standard box plots with median (twenty-fifth and seventy-fifth percentiles) and whiskers (at minimum and maximum values) of MRTF-A expression are depicted in the VL of the COPD patients and control subjects, outliers are represented by open dots. Statistical significance is represented as follows: n.s., non-significant among any of the study groups. Outliers are represented as open dots. (B) Standard box plots with median (twenty-fifth and seventy-fifth percentiles) and whiskers (at minimum and maximum values) of MRTF-B expression are depicted in the VL of the COPD patients and the control subjects, outliers are represented by open dots. Statistical significance is represented as follows: n.s., non-significant among any of the study groups. Outliers are represented as open dots. (C) Standard box plots with median (twenty-fifth and seventy-fifth percentiles) and whiskers (at minimum and maximum values) of SRF expression are depicted in the VL of the COPD patients and the control subjects, outliers are represented by open dots. Statistical significance is represented as follows: *P<0.05 between all patients with COPD and the control subjects; n.s., non-significant between both non-muscle-weakness and muscle-weakness COPD patients and the healthy controls. Outliers are represented as open dots.

Figure 5
Expression of myogenic transcription factors in the VL of patients with COPD with and without muscle weakness and healthy controls

(A) Standard box plots with median (twenty-fifth and seventy-fifth percentiles) and whiskers (at minimum and maximum values) of MRTF-A expression are depicted in the VL of the COPD patients and control subjects, outliers are represented by open dots. Statistical significance is represented as follows: n.s., non-significant among any of the study groups. Outliers are represented as open dots. (B) Standard box plots with median (twenty-fifth and seventy-fifth percentiles) and whiskers (at minimum and maximum values) of MRTF-B expression are depicted in the VL of the COPD patients and the control subjects, outliers are represented by open dots. Statistical significance is represented as follows: n.s., non-significant among any of the study groups. Outliers are represented as open dots. (C) Standard box plots with median (twenty-fifth and seventy-fifth percentiles) and whiskers (at minimum and maximum values) of SRF expression are depicted in the VL of the COPD patients and the control subjects, outliers are represented by open dots. Statistical significance is represented as follows: *P<0.05 between all patients with COPD and the control subjects; n.s., non-significant between both non-muscle-weakness and muscle-weakness COPD patients and the healthy controls. Outliers are represented as open dots.

Table 3
Transcription factors and sumoylation in VL of the study groups

Values are expressed as mean (S.D.).

Abbreviations:A, absorbance; a.u., arbitrary units; MyoD, myogenic differentiation; N, number of patients.

Statistical analyses: comparisons between all patients with COPD and the control subjects were assessed using the Student's ttest; comparisons among the two groups of COPD patients, non-muscle weakness and muscle-weakness and the healthy subjects were analysed using one-way ANOVA and Tukey's post-hoc analysis to adjust for multiple comparisons.

Statistical significance: **P≤0.01 between all patients with COPD and the control subjects; §P≤0.05 between either the non-muscle weakness or the muscle-weakness COPD patients and the control subjects; P≤0.05 between the muscle-weakness COPD patients and the non-muscle weakness patients.

COPD patients
Controls N=19All COPD patients N=41Non-muscle weakness N=16Muscle weakness N=25
Transcription factors involved in myogenesis 
MEF2C (A, a.u.) 0.253 (0.053) 0.215 (0.066) 0.226 (0.082) 0.206 (0.049) 
MEF2D (A, a.u.) 0.209 (0.093) 0.182 (0.090) 0.147 (0.079) 0.211 (0.090) 
MyoD (relative expression) 1 (0.58) 1.099 (0.59) 1.103 (0.58) 1.096 (0.62) 
Myogenin (relative expression) 1 (0.42) 1.002 (0.41) 1.012 (0.51) 0.995 (0.35) 
Hox-A11 (relative expression) 1 (0.43) 0.991 (0.46) 0.973 (0.47) 1 (0.46) 
YY1 (A, a.u.) 0.231 (0.049) 0.235 (0.068) 0.204 (0.048) 0.262 (0.072) 
Pax3 (relative expression) 1 (0.53) 0.671 (0.30)** 0.662 (0.32) 0.677 (0.30)§ 
Sumoylation markers 
SUMO-2 (relative expression) 1 (0.44) 0.815 (0.34) 0.894 (0.43) 0.773 (0.28) 
SUMO-3 (relative expression) 1 (0.34) 1.026 (0.35) 1.029 (0.36) 1.024 (0.35) 
COPD patients
Controls N=19All COPD patients N=41Non-muscle weakness N=16Muscle weakness N=25
Transcription factors involved in myogenesis 
MEF2C (A, a.u.) 0.253 (0.053) 0.215 (0.066) 0.226 (0.082) 0.206 (0.049) 
MEF2D (A, a.u.) 0.209 (0.093) 0.182 (0.090) 0.147 (0.079) 0.211 (0.090) 
MyoD (relative expression) 1 (0.58) 1.099 (0.59) 1.103 (0.58) 1.096 (0.62) 
Myogenin (relative expression) 1 (0.42) 1.002 (0.41) 1.012 (0.51) 0.995 (0.35) 
Hox-A11 (relative expression) 1 (0.43) 0.991 (0.46) 0.973 (0.47) 1 (0.46) 
YY1 (A, a.u.) 0.231 (0.049) 0.235 (0.068) 0.204 (0.048) 0.262 (0.072) 
Pax3 (relative expression) 1 (0.53) 0.671 (0.30)** 0.662 (0.32) 0.677 (0.30)§ 
Sumoylation markers 
SUMO-2 (relative expression) 1 (0.44) 0.815 (0.34) 0.894 (0.43) 0.773 (0.28) 
SUMO-3 (relative expression) 1 (0.34) 1.026 (0.35) 1.029 (0.36) 1.024 (0.35) 

Expression of SUMO

Expression levels of SUMO2 and SUMO3 did not differ between any of the study groups (Table 3).

Muscle structure

Fibre type composition

Compared with healthy controls, proportions of type I muscle fibres were significantly decreased whereas those of type II fibres were increased in VL of both groups of COPD patients together and when analysed separately (Table 4). Compared with healthy controls, the size of fast-twitch fibres, but not slow-twitch fibres, was lower in all COPD patients (16%) than in healthy controls and in muscle-weakness patients, fast-twitch fibre size was significantly reduced compared with both controls (25%) and non-muscle-weakness patients (22%; Table 4). No differences in type II fibre size were observed between non-muscle-weakness patients and healthy controls (Table 4). Among all COPD patients together, positive correlations were found between type II fibre sizes and SIRT-1 and BMI (r=0.491, P=0.033 and r=0.564, P=0.001 respectively). Furthermore, significant correlations were also observed between fast-twitch fibre sizes and proportions with BMI and FFMI respectively (r=0.494, P=0.0027 and r=−0.654, P=0.015 respectively) in muscle-weakness patients as a group.

Table 4
Fibre type composition in VL muscles of the study subjects

Values are expressed as mean (S.D.).

Abbreviations: CSA, cross-sectional area; N, number of patients.

Statistical analyses: comparisons between all patients with COPD and the control subjects were assessed using the Student's ttest; comparisons among the two groups of COPD patients, non-muscle weakness and muscle-weakness and the healthy subjects were analysed using one-way ANOVA and Tukey's post-hoc analysis to adjust for multiple comparisons.

Statistical significance: *P≤0.05, ***P≤0.001 between all patients with COPD and the control subjects; §P≤0.05, §§P≤0.01 between either the non-muscle-weakness or the muscle-weakness COPD patients and the control subjects; P≤0.05 between the muscle-weakness COPD patients and the non-muscle weakness patients.

COPD patients
Controls N=19All COPD patients N=41Non-muscle weakness N=16Muscle weakness N=25
Muscle fibre type composition 
Type I fibres, percentages 39 (6) 27 (10)*** 31 (10)§ 26 (9)§§ 
Type II fibres, percentages 61 (6) 73 (10)*** 69 (10)§ 74 (9)§§ 
Type I fibres, CSA (μm22698 (894) 2496 (855) 2618 (960) 2438 (817) 
Type II fibres, CSA (μm22915 (755) 2436 (847)* 2807 (910) 2188 (718)§¶ 
COPD patients
Controls N=19All COPD patients N=41Non-muscle weakness N=16Muscle weakness N=25
Muscle fibre type composition 
Type I fibres, percentages 39 (6) 27 (10)*** 31 (10)§ 26 (9)§§ 
Type II fibres, percentages 61 (6) 73 (10)*** 69 (10)§ 74 (9)§§ 
Type I fibres, CSA (μm22698 (894) 2496 (855) 2618 (960) 2438 (817) 
Type II fibres, CSA (μm22915 (755) 2436 (847)* 2807 (910) 2188 (718)§¶ 

DISCUSSION

In view of the reported findings, the study hypothesis has been confirmed; patients with advanced COPD and quadriceps weakness exhibit a differential profile of epigenetic mediators and events in their VL compared with patients without muscle weakness. Expression of miR-1, miR-206 and miR-27 were significantly up-regulated in the limb muscles of all COPD patients. Additionally, in the post-hoc analyses, the muscle-weakness patients also showed an up-regulation of the same miRNAs in their VL. Moreover, compared with healthy controls, levels of total acetyl-lysine proteins and histones and specific lysine-acetylated histone 3 were also increased in the limb muscles of all COPD patients together and the muscle-weakness patients when analysed separately. Compared with healthy controls, levels of HDAC3 and HDAC4 were lower in limb muscles of all COPD patients together and in muscle-weakness patients and those of SIRT-1 were decreased only in muscle-weakness patients. Interestingly, levels of several mediators of muscle mass maintenance and growth pathways known to be targeted by miR-1, miR-206 and miR-27 were also explored in the study; in the limb muscles of all COPD and muscle-weakness patients, HDAC4, IGF-1, myostatin and Pax3 expression was down-regulated, whereas follistatin levels were up-regulated.

It should be underscored that airway obstruction, functional signs of emphysema and exercise capacity were markedly impaired in patients with muscle weakness compared with those without muscle weakness. Other relevant findings in the study were that similar to previous data [33], muscle-weakness patients exhibited clear signs of systemic disease together with significant body composition impairment, loss of body weight and fast-twitch fibre atrophy irrespective of smoking history, since the proportions in active smokers were, indeed, significantly lower than in patients with no muscle weakness. Hence, it seems likely that in COPD, alterations in body and muscle fibre-type composition determine a differential phenotype characterized by prominent muscle weakness and signs of systemic disease regardless of age or smoking history. These results are also in line with previous findings [28,34], in which the size of type II fibres was consistently smaller in two distinct models of muscle wasting. Taken together, it could be hypothesized that fast-twitch, rather than slow-twitch, fibres seem to be more prone to degradation in those conditions. Interestingly, significant associations were also found between proportions and size of slow- and fast-twitch fibres and body composition parameters in muscle-weakness patients. These findings suggest that fibre-type composition and the size of fast-twitch fibres may rely on the nutritional status of the patients, as previously shown in other muscles [35].

As far as we are concerned, this is the first investigation reporting data on epigenetic regulation in the lower limb muscles of patients with advanced COPD and severe muscle weakness and atrophy. In addition to the classic transcription factors, muscle-specific miRNAs regulate muscle development and repair after injury by targeting different pathways [14,16,36]. Whereas miR-133 induces myoblast proliferation by inhibiting myotube formation [14,16,36,37], miR-1 and miR-206 promote cell differentiation and innervation [14,16,3638]. In the current study, expression of miR-1, miR-206 and miR-27a was up-regulated in the VL of all COPD patients as a group and in patients with severe muscle weakness. Moreover, several downstream targets of these epigenetic mediators have also been analysed in the limb muscles of the study patients. Specifically, levels of HDAC4 were decreased in the VL of all COPD patients together and in the muscle-weakness group, probably as a response to miR-1 up-regulation in the muscles of the same groups. Indeed, it has been previously demonstrated that miR-1 promotes myogenesis by targeting the transcriptional repression of HDAC4 gene expression in muscles [39]. Conversely, miR-133 induces myoblast proliferation by repressing SRF [39]. In the present study, SRF expression was up-regulated in muscles of all COPD patients together most probably as a result of the lack of differences in miR-133 expression between patients and healthy controls. Additionally, expression of miR-27 was shown to favour myogenic differentiation through induction of Pax3 down-regulation [40], as also shown to occur in the VL of all COPD patients and in those with severe muscle weakness in the current investigation. Altogether, these findings suggest that the biological control of muscle differentiation is relatively preserved in VL of patients with advanced COPD, whereas the ability to form new muscle (muscle proliferation signalling) would be hampered in those muscles.

Importantly, the findings reported herein are also in agreement with a previous report, in which an increase in miR-1 and a decrease in HDAC4 were demonstrated in VL of patients with mild COPD [21]. Nonetheless, in a previous investigation [16], limb muscles from patients with advanced COPD showed a significant reduction in miR-1 with a concomitant rise in HDAC4 levels [16]. Differences in the study models, biological methodologies, smoking history, which in the current study was similar between patients and controls in order to avoid this confounding factor, and physical activity could account for the discrepancies encountered between the two investigations [16]. It should also be mentioned that in the diaphragm muscle of patients with a wide range of airflow limitation and preserved body composition, miR-1 levels were also decreased whereas HDAC4 expression was increased [22]. Differences in the expression of different epigenetic mediators are also expected between limb and respiratory muscles in COPD.

Other relevant findings in the investigation were the down-regulation of IGF-1 in limb muscles of all COPD patients and in the muscle-weakness group, probably as a result of miR-1 up-regulation expression [41], whereas muscle Akt expression did not vary between patients and healthy controls. Collectively, these findings lead to the conclusion that muscle anabolism and growth signalling pathways are hindered in atrophied weak muscles of patients with advanced COPD. Interestingly, expression of follistatin was significantly increased, whereas that of myostatin was decreased in VL of all COPD patients and in those with severe muscle weakness. The functional implications of these specific findings are that a negative feedback loop may exist in the weak muscles of the COPD patients in order to prevent them from undergoing further atrophy. These results are in line with a previous report [28], in which myostatin protein levels did not differ between COPD patients and controls. Nevertheless, previous reports showed an increase in myostatin levels in both respiratory and limb muscles of COPD patients [4244]. Differences in study design, disease severity, muscle mass loss and body composition may account for the discrepancies between studies.

Transcription states are regulated by post-translational modifications of histones that are epigenetic markers for chromatin structure and function. Acetylation of lysine residues, which is a trigger for chromatin remodelling, regulates gene expression. Histone acetylation leads to gene transcription, whereas histone deacetylation is associated with gene silencing. Interestingly, acetylation of H3 takes place at several lysine residues in the histone tail that actively contributes to the dynamic regulation of gene expression of pathways mediating cellular proliferation, differentiation and death [45,46]. In the study, relevant novel findings are being reported in that a rise in the levels of total and histone lysine-acetylated proteins were observed in the limb muscles of all COPD patients and in those with severe muscle weakness. Specifically, acetylation of H3 was significantly increased in the muscles of the same groups of patients. Furthermore, significant inverse correlations were found between body composition parameters (BMI and FFMI) and quadriceps strength and levels of lysine-acetylated histones and those of H3. Increased levels of YY1 were seen in limb muscles of patients with muscle weakness. Interestingly, YY1 transcriptional activity may also modify histone acetylation levels, which in turn, inhibits muscle regeneration through transcriptional silencing of myofibrillar genes [47].

On the other hand, hyperacetylation of proteins in tissues may participate in the process of muscle wasting through several mechanisms, such as activation of transcription factors and nuclear cofactors, by rendering proteins more prone to catabolism by ubiquitin-ligase activity of several HATs and by dissociation of proteins from cellular chaperones [48]. Among several HATs, the nuclear cofactor p300 has been shown to regulate muscle differentiation and wasting in several experimental models [23,49]. In the current investigation, no differences were found in p300 expression levels among study subjects. Differences between study models and species may account for such discrepancies. Importantly, protein acetylation also relies on histone deacetylase activity. Again in experimental models of muscle wasting [23,49], levels of HDAC3, HDAC6 and SIRT-1 were shown to be decreased in muscles. In keeping with this, in the study, levels of HDAC3 and SIRT-1 were also down-regulated in the limb muscles of the muscle-weakness patients compared with healthy controls, whereas those of HDAC6 did not differ between patients and control subjects. Additionally, the significant inverse correlations found between either body composition or quadriceps strength and protein hyperacetylation support the conclusion that acetylation of lysine residues of proteins is likely to interfere with muscle function and mass maintenance, by enhancing protein breakdown, in COPD muscle weakness.

Sarcopenia and muscle wasting could be partly the result of the premature senescence of satellite cells, which ultimately would lead to decreased growth and poor repair potential. Interestingly, accumulation of SUMO ligases has been proposed to participate in the aetiology of premature senescence of primary myogenic cultured cells [50] and other models [51,52]. Furthermore, premature satellite cell senescence through SUMO ligases among other mechanisms also seems to underlie the pathophysiology of muscular dystrophies [25]. In the current study, expression levels of SUMO-2 and -3 did not differ between study groups, thus suggesting that premature senescence may not be a relevant mechanism of muscle weakness and atrophy in COPD.

Study limitations

A limitation has to do with the fact that smoking history was similar between the COPD groups of patients and the healthy controls. However, we purposely recruited non-COPD control subjects who had a similar smoking history in order to avoid potential biases inherent to differences in smoking loads among the study groups.

Another limitation is related to the use of sedentary individuals as the healthy controls. We reasoned that physical activity might largely influence the expression of the epigenetic events analysed herein. As the COPD patients were all sedentary, we decided to specifically recruit healthy sedentary individuals as the control group.

Finally, another potential limitation involves the descriptive nature of the study. However, the novelty of the results showing a differential pattern of expression of several epigenetic events in COPD patients with and without muscle weakness will be the basis for the design of interventional studies targeted to promote myogenesis and inhibit muscle wasting.

Physiological significance of the study findings and future research

Despite that the cross-sectional nature of the investigation does not allow the establishment of cause–effect relationships between the epigenetic mediators analysed in the study and muscle weakness, we believe it provides insight into epigenetic regulation of muscle growth and atrophy signalling in limb muscles of COPD patients with significant muscle weakness. Specifically, in the muscles of these patients, genes involved in muscle differentiation were predominantly regulated over genes involved in muscle proliferation. These findings have functional implications as proliferation of myoblasts from satellite cells is required for muscle regeneration and faulty regeneration further contributes to muscle atrophy in COPD [53]. Therefore, future research should aim to target mediators of muscle proliferation, while maintaining active muscle differentiation.

Other functional implications in the study are the increased levels of lysine-acetylation of both muscle proteins and histones observed in the VL of patients with severe muscle weakness. Further studies should focus on the elucidation of how histone and protein acetylation drive enhanced muscle catabolism and direct fast-twitch fibre atrophy. Furthermore, the potential beneficial effects of histone acetylation inhibitors should also be explored in future investigations.

Conclusions

In VL of severe COPD patients with muscle weakness and atrophy, epigenetic events seem to regulate muscle differentiation rather than proliferation, as well as muscle growth and atrophy signalling, probably as feedback mechanisms to prevent those muscles from further mass loss. Lysine-hyperacetylation of histones and other non-histone proteins may orchestrate enhanced protein catabolism in those muscles, contributing to further atrophy and weakness. These findings may help design novel therapeutic strategies (selective enhancers of miRNAs promoting myogenesis and acetylation inhibitors) to specifically target muscle weakness and atrophy as major comorbidities in severe COPD.

Abbreviations

     
  • Akt

    v-akt murine thymoma viral oncogene homologue 1

  •  
  • BMI

    body mass index

  •  
  • COPD

    chronic obstructive pulmonary disease

  •  
  • FFMI

    fat-free mass index

  •  
  • HAT

    histone acetyltransferase

  •  
  • HDAC

    histone deacetylase

  •  
  • Hox

    homoeobox protein

  •  
  • IGF

    insulin-like growth factor-1

  •  
  • MEF

    myocyte enhancer factor

  •  
  • MRTF

    myocardin-related transcription factor

  •  
  • Pax

    paired box

  •  
  • QMVC

    quadriceps maximal velocity contraction

  •  
  • RT

    reverse transcription

  •  
  • SIRT

    sirtuin

  •  
  • SRF

    serum reponse factor

  •  
  • SUMO

    small ubiquitin-related modifier

  •  
  • VL

    vastus lateralis

  •  
  • YY

    Yin Yang

AUTHOR CONTRIBUTION

Esther Barreiro conceptualized and designed the experiments. Planning of the experiments was carried out in the investigation by Esther Barreiro and Ester Puig-Vilanova. Acquisition of the reagents and materials required for the molecular biology experiments was performed by Ester Puig-Vilanova and Esther Barreiro. Full assessment of patients and healthy controls, recruitment of the candidate patients and controls according to the inclusion and exclusion criteria established in the investigation and sample collection from the study subjects were performed by Juana Martinez-Llorens, Pilar Ausin, Josep Roca and Joaquim Gea. Molecular biology analyses were done by Ester Puig-Vilanova and Esther Barreiro. Statistical analyses, data interpretation and presentation were performed by Ester Puig-Vilanova, Esther Barreiro and Joaquim Gea. Preparation of the manuscript figures and tables was done by Ester Puig-Vilanova and Esther Barreiro. Ester Puig-Vilanova, Joaquim Gea and Esther Barreiro drafted the manuscript and provided intellectual input. Manuscript writing final version was performed by Esther Barreiro. Esther Barreiro is the guarantor and is responsible for the overall content in the manuscript and writing.

The authors are grateful to Mireia Admetlló and Dr Sergi Pascual-Guardia for their help with patient reports and information and Mr Francisco Sánchez and Dr Carme Casadevall for their contribution to muscle structure and miRNA expression analyses. The authors are also very grateful to Mr Sergio Mojal for his continuous statistical advice and analyses of the study results.

FUNDING

This work was supported by the Centro de Investigacion Biomedica en Red [grant numbers FIS 11/02029, FIS 12/02534, SAF-2011-26908, 2009-SGR-393, SEPAR 2009, SEPAR 2014, FUCAP 2011 and FUCAP 2012]; the Marató TV3 [grant number MTV3-07-1010]; and the ERS COPD Research Award 2008 (to E.B.).

References

References
1
Miravitlles
 
M.
Soler-Cataluna
 
J.J.
Calle
 
M.
Molina
 
J.
Almagro
 
P.
Quintano
 
J.A.
Riesco
 
J.A.
Trigueros
 
J.A.
Pinera
 
P.
Simon
 
A.
, et al 
Spanish COPD Guidelines (GesEPOC): pharmacological treatment of stable COPD. Spanish Society of Pulmonology and Thoracic Surgery
Arch. Bronconeumol.
2012
, vol. 
48
 (pg. 
247
-
257
)
[PubMed]
2
Miravitlles
 
M.
Soler-Cataluna
 
J.J.
Calle
 
M.
Molina
 
J.
Almagro
 
P.
Antonio
 
Q.J.
Antonio
 
R.J.
Antonio
 
T.J.
Pinera
 
P.
Simon
 
A.
, et al 
Spanish guideline for COPD (GesEPOC). Update 2014
Arch. Bronconeumol.
2014
, vol. 
50
 
Suppl I
(pg. 
1
-
16
)
[PubMed]
3
Gayan-Ramirez
 
G.
Decramer
 
M.
 
Mechanisms of striated muscle dysfunction during acute exacerbations of COPD
J. Appl. Physiol.
2013
, vol. 
114
 (pg. 
1291
-
1299
)
[PubMed]
4
Gosselink
 
R.
Troosters
 
T.
Decramer
 
M.
 
Peripheral muscle weakness contributes to exercise limitation in COPD
Am. J. Respir. Crit. Care Med.
1996
, vol. 
153
 (pg. 
976
-
980
)
[PubMed]
5
Seymour
 
J.M.
Spruit
 
M.A.
Hopkinson
 
N.S.
Natanek
 
S.A.
Man
 
W.D.
Jackson
 
A.
Gosker
 
H.R.
Schols
 
A.M.
Moxham
 
J.
Polkey
 
M.I.
, et al 
The prevalence of quadriceps weakness in COPD and the relationship with disease severity
Eur. Respir. J.
2010
, vol. 
36
 (pg. 
81
-
88
)
[PubMed]
6
Marquis
 
K.
Debigare
 
R.
Lacasse
 
Y.
LeBlanc
 
P.
Jobin
 
J.
Carrier
 
G.
Maltais
 
F.
 
Midthigh muscle cross-sectional area is a better predictor of mortality than body mass index in patients with chronic obstructive pulmonary disease
Am. J. Respir. Crit. Care Med.
2002
, vol. 
166
 (pg. 
809
-
813
)
[PubMed]
7
Swallow
 
E.B.
Reyes
 
D.
Hopkinson
 
N.S.
Man
 
W.D.
Porcher
 
R.
Cetti
 
E.J.
Moore
 
A.J.
Moxham
 
J.
Polkey
 
M.I.
 
Quadriceps strength predicts mortality in patients with moderate to severe chronic obstructive pulmonary disease
Thorax
2007
, vol. 
62
 (pg. 
115
-
120
)
[PubMed]
8
Maltais
 
F.
Decramer
 
M.
Casaburi
 
R.
Barreiro
 
E.
Burelle
 
Y.
Debigare
 
R.
Dekhuijzen
 
P.N.
Franssen
 
F.
Gayan-Ramirez
 
G.
Gea
 
J.
, et al 
An official american thoracic Society/European respiratory society statement: update on limb muscle dysfunction in chronic obstructive pulmonary disease
Am. J. Respir. Crit. Care Med.
2014
, vol. 
189
 (pg. 
e15
-
e62
)
[PubMed]
9
Gea
 
J.
Agusti
 
A.
Roca
 
J.
 
Pathophysiology of muscle dysfunction in COPD
J. Appl. Physiol.
2013
, vol. 
114
 (pg. 
1222
-
1234
)
[PubMed]
10
Mantilla
 
C.B.
Sieck
 
G.C.
 
Neuromotor control in chronic obstructive pulmonary disease
J. Appl. Physiol.
2013
, vol. 
114
 (pg. 
1246
-
1252
)
[PubMed]
11
Levine
 
S.
Bashir
 
M.H.
Clanton
 
T.L.
Powers
 
S.K.
Singhal
 
S.
 
COPD elicits remodeling of the diaphragm and vastus lateralis muscles in humans
J. Appl. Physiol.
2013
, vol. 
114
 (pg. 
1235
-
1245
)
[PubMed]
12
Remels
 
A.H.
Gosker
 
H.R.
Langen
 
R.C.
Schols
 
A.M.
 
The mechanisms of cachexia underlying muscle dysfunction in COPD
J. Appl. Physiol.
2013
, vol. 
114
 (pg. 
1253
-
1262
)
[PubMed]
13
Puente-Maestu
 
L.
Lazaro
 
A.
Humanes
 
B.
 
Metabolic derangements in COPD muscle dysfunction
J. Appl. Physiol.
2013
, vol. 
114
 (pg. 
1282
-
1290
)
[PubMed]
14
Barreiro
 
E.
Sznajder
 
J.I.
 
Epigenetic regulation of muscle phenotype and adaptation: a potential role in COPD muscle dysfunction
J. Appl. Physiol.
2013
, vol. 
114
 (pg. 
1263
-
1272
)
[PubMed]
15
Donaldson
 
A.
Natanek
 
S.A.
Lewis
 
A.
Man
 
W.D.
Hopkinson
 
N.S.
Polkey
 
M.I.
Kemp
 
P.R.
 
Increased skeletal muscle-specific microRNA in the blood of patients with COPD
Thorax
2013
, vol. 
68
 (pg. 
1140
-
1149
)
[PubMed]
16
Lewis
 
A.
Riddoch-Contreras
 
J.
Natanek
 
S.A.
Donaldson
 
A.
Man
 
W.D.
Moxham
 
J.
Hopkinson
 
N.S.
Polkey
 
M.I.
Kemp
 
P.R.
 
Downregulation of the serum response factor/miR-1 axis in the quadriceps of patients with COPD
Thorax
2012
, vol. 
67
 (pg. 
26
-
34
)
[PubMed]
17
McCarthy
 
J.J.
Esser
 
K.A.
 
MicroRNA-1 and microRNA-133a expression are decreased during skeletal muscle hypertrophy
J. Appl. Physiol.
2007
, vol. 
102
 (pg. 
306
-
313
)
[PubMed]
18
McCarthy
 
J.J.
Esser
 
K.A.
Peterson
 
C.A.
Dupont-Versteegden
 
E.E.
 
Evidence of MyomiR network regulation of beta-myosin heavy chain gene expression during skeletal muscle atrophy
Physiol. Genomics
2009
, vol. 
39
 (pg. 
219
-
226
)
[PubMed]
19
Allen
 
D.L.
Bandstra
 
E.R.
Harrison
 
B.C.
Thorng
 
S.
Stodieck
 
L.S.
Kostenuik
 
P.J.
Morony
 
S.
Lacey
 
D.L.
Hammond
 
T.G.
Leinwand
 
L.L.
, et al 
Effects of spaceflight on murine skeletal muscle gene expression
J. Appl. Physiol.
2009
, vol. 
106
 (pg. 
582
-
595
)
[PubMed]
20
Natanek
 
S.A.
Riddoch-Contreras
 
J.
Marsh
 
G.S.
Hopkinson
 
N.S.
Man
 
W.D.
Moxham
 
J.
Polkey
 
M.I.
Kemp
 
P.R.
 
Yin Yang 1 expression and localisation in quadriceps muscle in COPD
Arch. Bronconeumol.
2011
, vol. 
47
 (pg. 
296
-
302
)
[PubMed]
21
Puig-Vilanova
 
E.
Ausin
 
P.
Martinez-Llorens
 
J.
Gea
 
J.
Barreiro
 
E.
 
Do epigenetic events take place in the vastus lateralis of patients with mild chronic obstructive pulmonary disease?
PLoS One
2014
, vol. 
9
 pg. 
e102296
 
[PubMed]
22
Puig-Vilanova
 
E.
Aguilo
 
R.
Rodriguez-Fuster
 
A.
Martinez-Llorens
 
J.
Gea
 
J.
Barreiro
 
E.
 
Epigenetic mechanisms in respiratory muscle dysfunction of patients with chronic obstructive pulmonary disease
PLoS One
2014
, vol. 
9
 pg. 
e111514
 
[PubMed]
23
Alamdari
 
N.
Smith
 
I.J.
Aversa
 
Z.
Hasselgren
 
P.O.
 
Sepsis and glucocorticoids upregulate p300 and downregulate HDAC6 expression and activity in skeletal muscle
Am. J. Physiol. Regul. Integr. Comp. Physiol.
2010
, vol. 
299
 (pg. 
R509
-
R520
)
[PubMed]
24
Seigneurin-Berny
 
D.
Verdel
 
A.
Curtet
 
S.
Lemercier
 
C.
Garin
 
J.
Rousseaux
 
S.
Khochbin
 
S.
 
Identification of components of the murine histone deacetylase 6 complex: link between acetylation and ubiquitination signaling pathways
Mol. Cell. Biol.
2001
, vol. 
21
 (pg. 
8035
-
8044
)
[PubMed]
25
Kudryashova
 
E.
Kramerova
 
I.
Spencer
 
M.J.
 
Satellite cell senescence underlies myopathy in a mouse model of limb-girdle muscular dystrophy 2H
J. Clin. Invest.
2012
, vol. 
122
 (pg. 
1764
-
1776
)
[PubMed]
26
Miravitlles
 
M.
Calle
 
M.
Soler-Cataluna
 
J.J.
 
Clinical phenotypes of COPD: identification, definition and implications for guidelines
Arch. Bronconeumol.
2012
, vol. 
48
 (pg. 
86
-
98
)
[PubMed]
27
Barreiro
 
E.
Peinado
 
V.I.
Galdiz
 
J.B.
Ferrer
 
E.
Marin-Corral
 
J.
Sanchez
 
F.
Gea
 
J.
Barbera
 
J.A.
 
Cigarette smoke-induced oxidative stress: a role in chronic obstructive pulmonary disease skeletal muscle dysfunction
Am. J. Respir. Crit. Care Med.
2010
, vol. 
182
 (pg. 
477
-
488
)
[PubMed]
28
Fermoselle
 
C.
Rabinovich
 
R.
Ausin
 
P.
Puig-Vilanova
 
E.
Coronell
 
C.
Sanchez
 
F.
Roca
 
J.
Gea
 
J.
Barreiro
 
E.
 
Does oxidative stress modulate limb muscle atrophy in severe COPD patients?
Eur. Respir. J.
2012
, vol. 
40
 (pg. 
851
-
862
)
[PubMed]
29
Livak
 
K.J.
Schmittgen
 
T.D.
 
Analysis of relative gene expression data using real-time quantitative PCR and the 2(-delta delta C(T)) method
Methods
2001
, vol. 
25
 (pg. 
402
-
408
)
[PubMed]
30
Schulz
 
M.D.
Atay
 
C.
Heringer
 
J.
Romrig
 
F.K.
Schwitalla
 
S.
Aydin
 
B.
Ziegler
 
P.K.
Varga
 
J.
Reindl
 
W.
Pommerenke
 
C.
, et al 
High-fat-diet-mediated dysbiosis promotes intestinal carcinogenesis independently of obesity
Nature
2014
, vol. 
514
 (pg. 
508
-
512
)
[PubMed]
31
Habib
 
O.
Habib
 
G.
Choi
 
H.W.
Hong
 
K.S.
Do
 
J.T.
Moon
 
S.H.
Chung
 
H.M.
 
An improved method for the derivation of high quality iPSCs in the absence of c-Myc
Exp. Cell Res.
2013
, vol. 
319
 (pg. 
3190
-
3200
)
[PubMed]
32
Hernandez-Jimenez
 
M.
Hurtado
 
O.
Cuartero
 
M.I.
Ballesteros
 
I.
Moraga
 
A.
Pradillo
 
J.M.
McBurney
 
M.W.
Lizasoain
 
I.
Moro
 
M.A.
 
Silent information regulator 1 protects the brain against cerebral ischemic damage
Stroke
2013
, vol. 
44
 (pg. 
2333
-
2337
)
[PubMed]
33
Op den Kamp
 
C.M.
Langen
 
R.C.
Minnaard
 
R.
Kelders
 
M.C.
Snepvangers
 
F.J.
Hesselink
 
M.K.
Dingemans
 
A.C.
Schols
 
A.M.
 
Pre-cachexia in patients with stages I-III non-small cell lung cancer: systemic inflammation and functional impairment without activation of skeletal muscle ubiquitin proteasome system
Lung Cancer
2012
, vol. 
76
 (pg. 
112
-
117
)
[PubMed]
34
Marin-Corral
 
J.
Fontes
 
C.C.
Pascual-Guardia
 
S.
Sanchez
 
F.
Olivan
 
M.
Argiles
 
J.M.
Busquets
 
S.
Lopez-Soriano
 
F.J.
Barreiro
 
E.
 
Redox balance and carbonylated proteins in limb and heart muscles of cachectic rats
Antioxid. Redox Signal.
2010
, vol. 
12
 (pg. 
365
-
380
)
[PubMed]
35
Kelsen
 
S.G.
Ference
 
M.
Kapoor
 
S.
 
Effects of prolonged undernutrition on structure and function of the diaphragm
J. Appl. Physiol.
1985
, vol. 
58
 (pg. 
1354
-
1359
)
[PubMed]
36
Perdiguero
 
E.
Sousa-Victor
 
P.
Ballestar
 
E.
Munoz-Canoves
 
P.
 
Epigenetic regulation of myogenesis
Epigenetics
2009
, vol. 
4
 (pg. 
541
-
550
)
[PubMed]
37
Barreiro
 
E.
Gea
 
J.
 
Epigenetics and muscle dysfunction in chronic obstructive pulmonary disease
Transl. Res.
2015
, vol. 
165
 (pg. 
61
-
73
)
[PubMed]
38
Nakajima
 
N.
Takahashi
 
T.
Kitamura
 
R.
Isodono
 
K.
Asada
 
S.
Ueyama
 
T.
Matsubara
 
H.
Oh
 
H.
 
MicroRNA-1 facilitates skeletal myogenic differentiation without affecting osteoblastic and adipogenic differentiation
Biochem. Biophys. Res. Commun.
2006
, vol. 
350
 (pg. 
1006
-
1012
)
[PubMed]
39
Chen
 
J.F.
Mandel
 
E.M.
Thomson
 
J.M.
Wu
 
Q.
Callis
 
T.E.
Hammond
 
S.M.
Conlon
 
F.L.
Wang
 
D.Z.
 
The role of microRNA-1 and microRNA-133 in skeletal muscle proliferation and differentiation
Nat. Genet.
2006
, vol. 
38
 (pg. 
228
-
233
)
[PubMed]
40
Crist
 
C.G.
Montarras
 
D.
Pallafacchina
 
G.
Rocancourt
 
D.
Cumano
 
A.
Conway
 
S.J.
Buckingham
 
M.
 
Muscle stem cell behavior is modified by microRNA-27 regulation of Pax3 expression
Proc. Natl. Acad. Sci. U.S.A.
2009
, vol. 
106
 (pg. 
13383
-
13387
)
[PubMed]
41
Elia
 
L.
Contu
 
R.
Quintavalle
 
M.
Varrone
 
F.
Chimenti
 
C.
Russo
 
M.A.
Cimino
 
V.
De
 
M.L.
Frustaci
 
A.
Catalucci
 
D.
, et al 
Reciprocal regulation of microRNA-1 and insulin-like growth factor-1 signal transduction cascade in cardiac and skeletal muscle in physiological and pathological conditions
Circulation
2009
, vol. 
120
 (pg. 
2377
-
2385
)
[PubMed]
42
Man
 
C.
Natanek
 
S.A.
Riddoch-Contreras
 
J.
Lewis
 
A.
Marsh
 
G.S.
Kemp
 
P.R.
Polkey
 
M.I.
 
Quadriceps myostatin expression in COPD
Eur. Respir. J.
2010
, vol. 
36
 (pg. 
686
-
688
)
[PubMed]
43
Testelmans
 
D.
Crul
 
T.
Maes
 
K.
Agten
 
A.
Crombach
 
M.
Decramer
 
M.
Gayan-Ramirez
 
G.
 
Atrophy and hypertrophy signalling in the diaphragm of patients with COPD
Eur. Respir. J.
2010
, vol. 
35
 (pg. 
549
-
556
)
[PubMed]
44
Vogiatzis
 
I.
Simoes
 
D.C.
Stratakos
 
G.
Kourepini
 
E.
Terzis
 
G.
Manta
 
P.
Athanasopoulos
 
D.
Roussos
 
C.
Wagner
 
P.D.
Zakynthinos
 
S.
 
Effect of pulmonary rehabilitation on muscle remodelling in cachectic patients with COPD
Eur. Respir. J.
2010
, vol. 
36
 (pg. 
301
-
310
)
[PubMed]
45
Suka
 
N.
Suka
 
Y.
Carmen
 
A.A.
Wu
 
J.
Grunstein
 
M.
 
Highly specific antibodies determine histone acetylation site usage in yeast heterochromatin and euchromatin
Mol. Cell
2001
, vol. 
8
 (pg. 
473
-
479
)
[PubMed]
46
Vempati
 
R.K.
Jayani
 
R.S.
Notani
 
D.
Sengupta
 
A.
Galande
 
S.
Haldar
 
D.
 
p300-mediated acetylation of histone H3 lysine 56 functions in DNA damage response in mammals
J. Biol. Chem.
2010
, vol. 
285
 (pg. 
28553
-
28564
)
[PubMed]
47
Wang
 
H.
Hertlein
 
E.
Bakkar
 
N.
Sun
 
H.
Acharyya
 
S.
Wang
 
J.
Carathers
 
M.
Davuluri
 
R.
Guttridge
 
D.C.
 
NF-kappaB regulation of YY1 inhibits skeletal myogenesis through transcriptional silencing of myofibrillar genes
Mol. Cell Biol.
2007
, vol. 
27
 (pg. 
4374
-
4387
)
[PubMed]
48
Alamdari
 
N.
Aversa
 
Z.
Castillero
 
E.
Hasselgren
 
P.O.
 
Acetylation and deacetylation–novel factors in muscle wasting
Metabolism
2013
, vol. 
62
 (pg. 
1
-
11
)
[PubMed]
49
Sadoul
 
K.
Boyault
 
C.
Pabion
 
M.
Khochbin
 
S.
 
Regulation of protein turnover by acetyltransferases and deacetylases
Biochimie
2008
, vol. 
90
 (pg. 
306
-
312
)
[PubMed]
50
Andreou
 
A.M.
Tavernarakis
 
N.
 
SUMOylation and cell signalling
Biotechnol. J.
2009
, vol. 
4
 (pg. 
1740
-
1752
)
[PubMed]
51
Bischof
 
O.
Schwamborn
 
K.
Martin
 
N.
Werner
 
A.
Sustmann
 
C.
Grosschedl
 
R.
Dejean
 
A.
 
The E3 SUMO ligase PIASy is a regulator of cellular senescence and apoptosis
Mol. Cell
2006
, vol. 
22
 (pg. 
783
-
794
)
[PubMed]
52
Bischof
 
O.
Dejean
 
A.
 
SUMO is growing senescent
Cell Cycle
2007
, vol. 
6
 (pg. 
677
-
681
)
[PubMed]
53
Theriault
 
M.E.
Pare
 
M.E.
Lemire
 
B.B.
Maltais
 
F.
Debigare
 
R.
 
Regenerative defect in vastus lateralis muscle of patients with chronic obstructive pulmonary disease
Respir. Res.
2014
, vol. 
15
 pg. 
35
 
[PubMed]

Supplementary data